Many youth and young adults are exposed to marijuana and alcohol-related content via social media. Because substance use tracks with perceptions of harm, this promotion of marijuana and alcohol is worrisome for a potential increase in adolescent and young adult substance use. Given its irreversible penetration among young people, social media can be used to help prevent hazardous substance use behaviors. Health organizations can deliver messages to target populations via social media. However, such content might be offset by conflicting posts. In this context, Aim 1 of our study will unravel the nature, extent, and engagement patterns of messages about substance use on a popular social media outlet (Twitter) and describe the engagement patterns of networks Tweeting pro- and anti-substance use messages. Here, we will assess the engagement/influence of marijuana and alcohol Tweets, monitor their temporal trends, and dissect their sentiment/content. In addition, we will assess the Tweets and engagement patterns from popular Twitter handles that daily Tweet about substance use, and study the demographics of people following these handles. We will use social network analytic methods to study the structure and interconnectedness of Twitter users Tweeting about marijuana use, alcohol use, or both substances, and the characteristics associated with being influential players in these networks. It is not enough to study the substance use chatter on Twitter; it is critical to know how exposure is related to substance use behaviors. To assess the behavioral fidelity of our Twitter findings, we will survey 3,000 young adult social media users to study the implications of substance use content exposure across multiple social media platforms (e.g., Twitter, Facebook and YouTube). The Theory of Planned Behavior (TPB) has been used to predict substance use behaviors and will guide our analysis as we underscore how social media messages are associated with individuals' marijuana and alcohol use behaviors and their attitudes/social norms. Results will facilitate more strategic interventions that exploit the content and dynamics of marijuana and alcohol-related information exchanges on social media, and will aid in the tailoring and targeting of both online and offline prevention methods.